delivered the opinion of the Court.

Section 5 of the Voting Rights Act of 1965 1 prohibits *132a State or political subdivision subject to § 4 of the Act2 from enforcing “any voting qualification or prerequisite to voting, or standard, practice, or procedure with re*133spect to voting different from that in force or effect on November 1, 1964,” unless it has obtained a declaratory-judgment from the District Court for the District of Columbia that such change “does not have the purpose and will not have the effect of denying or abridging the right to vote on account of race or color” or has submitted the proposed change to the Attorney General and the Attorney General has not objected to it. The constitutionality of this procedure was upheld in South Carolina v. Katzenbach, 383 U. S. 301, and it is now well established that § 5 is applicable when a State or political subdivision adopts a legislative reapportionment plan. Allen v. State Board of Elections, 393 U. S. 544; Georgia v. United States, 411 U. S. 526.

The city of New Orleans brought this suit under § 5 seeking a judgment declaring that a reapportionment of New Orleans’ councilmanic districts did not have the purpose or effect of denying or abridging the right to vote on account of race or color.3 The District Court *134entered a judgment of dismissal, holding that the new reapportionment plan would have the effect of abridging the voting rights of New Orleans’ Negro citizens. 374 F. Supp. 363. The city appealed the judgment to this Court, claiming that the District Court used an incorrect standard in assessing the effect of the reapportionment in this § 5 suit. We noted probable jurisdiction of the appeal. 419 U. S. 822.

I

New Orleans is a city of almost 600,000 people. Some 55% of that population is white and the remaining 45% is Negro. Some 65% of the registered voters are white, and the remaining 35% are Negro.4 In 1954, New Orleans adopted a mayor-council form of government. Since that time the municipal charter has provided that the city council is to consist of seven members, one to be elected from each of five councilmanic districts, and two to be elected by the voters of the city at large. The 1954 charter also requires an adjustment of the boundaries of the five single-member councilmanic districts following each decennial census to reflect population shifts among the districts.

*135In 1961, the city council redistricted the city based on the 1960 census figures. That reapportionment plan established four districts that stretched from the edge of Lake Pontchartrain on the north side of the city to the Mississippi River on the city’s south side. The fifth district was wedge shaped and encompassed the city’s downtown area. In one of these councilmanic districts, Negroes constituted a majority’ of the population, but only about half of the registered voters. In the other four districts white voters clearly outnumbered Negro voters. No Negro was elected to the New Orleans City Council during the decade from 1960 to 1970.

After receipt of the 1970 census figures the city council adopted a reapportionment plan (Plan I) that continued the basic north-to-south pattern of councilmanic districts combined with a wedge-shaped, downtown district. Under Plan I Negroes constituted a majority of the population in two districts, but they did not make up a majority of registered voters in any district. The largest percentage of Negro voters in a single district under Plan I was 45.2%. When the city submitted Plan I to the Attorney General pursuant to § 5, he objected to it, stating that it appeared to “dilute black voting strength by combining a number of black voters with a larger number of white voters in each of the five districts.” He also expressed the view that “the district lines [were not] drawn as they [were] because of any compelling governmental need” and that the district lines did “not reflect numeric population configurations or considerations of district compactness or regularity of shape.”

Even before the Attorney General objected to Plan I, the city authorities had commenced work on a second plan — Plan II.5 That plan followed the general north-*136to-south districting pattern common to the 1961 apportionment and Plan I.6 It produced Negro population majorities in two districts and a Negro voter majority (52.6%) in one district. When Plan II was submitted to the Attorney General, he posed the same objections to it that he had raised to Plan I. In addition, he noted that “the predominantly black neighborhoods in the city are located generally in an east to west progression,” and pointed out that the use of north-to-south districts in such a situation almost inevitably would have the effect of diluting the maximum potential impact of the Negro vote. Following the rejection by the Attorney General of Plan II, the city brought this declaratory judgment action in the United States District Court for the District of Columbia.

The District Court concluded that Plan II would have the effect of abridging the right to vote on account of race or color.7 It calculated that if Negroes could elect city councilmen in proportion to their share of the city’s registered voters, they would be able to choose 2.42 of the city’s seven councilmen, and, if in proportion to their share of the city’s population, to choose 3.15 councilmen.8 But under Plan II the District Court concluded *137that, since New Orleans’ elections had been marked by bloc voting along racial lines, Negroes would probably be able to elect only one councilman — the candidate from the one councilmanic district in which a majority of the voters were Negroes. This difference between mathematical potential and predicted, reality was such that “the burden in the case at bar was at least to demonstrate that nothing but the redistricting proposed by Plan II was feasible.” 374 F. Supp., at 393. The court concluded that “[t]he City has not made that sort of demonstration; indeed, it was conceded at trial that neither that plan nor any of its variations was the City’s sole available alternative.” Ibid.9

As a separate and independent ground for rejecting Plan II, the District Court held that the failure of the plan to alter the city charter provision establishing two at-large seats had the effect in itself of “abridging the right to vote ... on account of race or color.” As the court put it: “[T]he City has not supported the choice of at-large elections by any consideration which would sat*138isfy the standard of compelling governmental interest, or the need to demonstrate the improbability of its realization through the use of single-member districts. These evaluations compel the conclusion that the feature of the city’s electoral scheme by which two councilmen are selected at large has the effect of impermissibly minimizing the vote of its black citizens; and the further conclusion that for this additional reason the city’s redistricting plan does not pass muster.” Id., at 402. (Footnotes omitted.)

The District Court therefore refused to allow Plan II to go into effect. As a result there have been no coun-cilmanic elections in New Orleans since 1970, and the councilmen elected at that time (or their appointed successors) have remained in office ever since.

II

A

The appellants urge, and the United States on reargument of this case has conceded, that the District Court was mistaken in holding that Plan II could be rejected under § 5 solely because it did not eliminate the two at-large councilmanic seats that had existed since 1954. The appellants and the United States are correct in their interpretation of the statute in this regard.

The language of § 5 clearly provides that it applies only to proposed changes in voting procedures. “[Discriminatory practices . . . instituted prior to November 1964 . . . are not subject to the requirement of pre-clearance [under § 5].” U. S. Commission on Civil Rights, The Voting Rights Act: Ten Years After, p. 347. The ordinance that adopted Plan II made no reference to the at-large councilmanic seats. Indeed, since those seats had been established in 1954 by the city charter, an ordinance could not have altered them; any change in *139the charter would have required approval by the city’s voters. The at-large seats, having existed without change since 1954, were not subject to review in this proceeding under § 5.10

B

The principal argument made by the appellants in this Court is that the District Court erred in concluding that the makeup of the five geographic council-manic districts under Plan II would have the effect of abridging voting rights on account of race or color. In evaluating this claim it is important to note at the outset that the question is not one of constitutional law, but of statutory construction.11 A determination of when a legislative reapportionment has “the effect of denying or abridging the right to vote on account of race or color,” must depend, therefore, upon the intent of *140Congress in enacting the Voting Rights Act and specifically § 5.

The legislative history reveals that the basic purpose of Congress in enacting the Voting Rights Act was “to rid the country of racial discrimination in voting.” South Carolina v. Katzenbach, 383 U. S., at 315. Section 5 was intended to play an important role in achieving that goal:

“Section 5 was a response to a common practice in some jurisdictions of staying one step ahead of the federal courts by passing new discriminatory voting laws as soon as the old ones had'been struck down. That practice had been possible because each new law remained in effect until the Justice Department or private plaintiffs were able to sustain the burden of proving that the new law, too, was discriminatory. . . . Congress therefore decided, as the Supreme Court held it could, ‘to shift the advantage of time and inertia from the perpetrators of the evil to its victim/ by ‘freezing election procedures in the covered areas unless the changes can be shown to be nondiscriminatory.’ ” H. R. Rep. No. 94-196, pp. 57-58. (Footnotes omitted.)

See also H. R. Rep. No. 439, 89th Cong., 1st Sess., 9-11, 26; S. Rep. No. 162, 89th Cong., 1st Sess., pt. 3, pp. 6-9, 24; H. R. Rep. No. 91-397, pp. 6-8; H. R. Rep. No. 94-196, pp. 8-11, 57-60; S. Rep. No. 94-295, pp. 15-19; South Carolina v. Katzenbach, supra, at 335.

By prohibiting the enforcement of a voting-procedure change until it has been demonstrated to the United States Department of Justice or to a three-judge federal court that the change does not have a discriminatory effect, Congress desired to prevent States from “undo-ting] or defeat [ing] the rights recently won” by Negroes. H. R. Rep. No. 91-397, p. 8. Section 5 was intended *141“to insure that [the gains thus far achieved in minority political participation] shall not be destroyed through new [discriminatory] procedures and techniques.” S. Rep. No. 94-295, p. 19.

When it adopted a 7-year extension of the Voting Rights Act in 1975, Congress explicitly stated that “the standard [under § 5] can only be fully satisfied by determining on the basis of the facts found by the Attorney General [or the District Court] to be true whether the ability of minority groups to participate in the political process and to elect their choices to office is augmented, diminished, or not affected by the change affecting voting . . . .” H. R. Rep. No. 94-196, p. 60 (emphasis added) ,12 In other words the purpose of § 5 has always been to insure that no voting-procedure changes would be made that would lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise.

It is thus apparent that a legislative reapportionment that enhances the position of racial minorities with respect to their effective exercise of the electoral franchise can hardly have the “effect” of diluting or abridging the right to vote on account of race within the meaning of § 5. We conclude, therefore, that such an ameliorative new legislative apportionment cannot violate § 5 unless the new apportionment itself so discriminates on the basis of race or color as to violate the Constitution.

The application of this standard to the facts of the present case is straightforward. Under the apportionment of 1961 none of the five councilmanic districts had a clear Negro majority of registered voters, and no Negro *142has been elected to the New Orleans City Council while that apportionment system has been in effect. Under Plan II, by contrast, Negroes will constitute a majority of the population in two of the five districts and a clear majority of the registered voters in one of them. Thus, there is every reason to predict, upon the District Court’s hypothesis of bloc voting, that at least one and perhaps two Negroes may well be elected to the council under Plan II.13 It was therefore error for the District Court to conclude that Plan II “will. . . have the effect of denying or abridging the right to vote on account of race or color” within the meaning of § 5 of the Voting Rights Act.14

*143Accordingly, the judgment of the District Court is vacated, and the case is remanded to that court for further proceedings consistent with this opinion.

It is so ordered.

Mr. Justice Stevens took no part in the consideration or decision of this case.

dissenting.

With Mr. Justice Marshall, I cannot agree that § 5 of the Voting Rights Act of 1965 reaches only those changes in election procedures that are more burdensome to the complaining minority than pre-existing procedures. As I understand § 5, the validity of any procedural change otherwise within the reach of the section must be determined under the statutory standard — whether the proposed legislation has the purpose or effect of abridging or denying the right to vote based on race or color.

This statutory standard is to be applied here in light of the District Court’s findings, which are supported by the evidence and are not now questioned by the Court. The findings were that the nominating process in New Orleans’ councilmanic elections is subject to majority vote and “anti-single-shot” rules and that there is a history of bloc racial voting in New Orleans, the predictable result being that no Negro candidate will win in any district in which his race is in the minority. In my view, where these facts exist, combined with a segregated residential pattern, § 5 is not satisfied unless, to the extent practicable, the new electoral districts afford the Negro minority the opportunity to achieve legislative representation roughly proportional to the Negro population *144in the community. Here, with a seven-member city council, the black minority constituting approximately 45% of the population of New Orleans, would be entitled under § 5, as I construe it, to the opportunity of electing at least three city councilmen — more than provided by the plan at issue here.

Bloc racial voting is an unfortunate phenomenon, but we are repeatedly faced with the findings of knowledgeable district courts that it is a fact of life. Where it exists, most often the result is that neither white nor black can be elected from a district in which his race is in the minority. As I see it, Congress has the power to minimize the effects of racial voting, particularly where it occurs in the context of other electoral rules operating to muffle the political potential of the minority. I am also satisfied that § 5 was aimed at this end, among others, and should be so construed and applied. See City of Richmond v. United States, 422 U. S. 358, 370-372 (1975).

Minimizing the exclusionary effects of racial voting is possible here because whites and blacks are not scattered evenly throughout the city; to á great extent, each race is concentrated in identifiable areas of New Orleans. But like bloc voting by race, this too is a fact of life, well known to those responsible for drawing electoral district lines. These lawmakers are quite aware that the districts they create will have a white or a black majority; and with each new district comes the unavoidable choice as to the racial composition of the district. It is here that § 5 intervenes to control these choices to the extent necessary to afford the minority the opportunity of achieving fair representation in the legislative body in question.

Applying § 5 in this way would at times require the drawing of district lines based on race; but Congress has this power where deliberate discrimination at the polls *145and the relevant electoral laws and customs have effectively foreclosed Negroes from enjoying a modicum of fair representation in the city council or other legislative body.

Since Plan II at issue in this case falls short of satisfying § 5 and since I agree with Mr. Justice Marshall that the city has failed to present sufficiently substantial justifications for its proposal, I respectfully dissent and would affirm the judgment of the District Court.

with whom Mr. Justice Brennan joins, dissenting.

Over the past 10 years the Court has, again and again, read the jurisdiction of § 5 of the Voting Rights Act of 1965, 79 Stat. 439, as amended, 89 Stat. 402, 404, 42 U. S. C. § 1973c (1970 ed., Supp. V), expansively so as “to give the Act the broadest possible scope” and to reach “any state enactment which altered the election law of a covered State in even a minor way.” Allen v. State Board of Elections, 393 U. S. 544, 567, 566 (1969). See also Georgia v. United States, 411 U. S. 526 (1973); Perkins v. Matthews, 400 U. S. 379 (1971); South Carolina v. Katzenbach, 383 U. S. 301 (1966). While we have settled the contours of § 5’s jurisdiction, however, we have yet to devote much attention to defining § 5’s substantive force within those bounds. Thus, we are faced today for the first time with the question of § 5’s substantive application to a redistricting plan. Essentially, we must answer one question: When does a redistricting plan have the effect of “abridging” the right to vote on account of race or color?

The Court never answers this question. Instead, it produces a convoluted construction of the statute that transforms the single question suggested by § 5 into three questions, and then provides precious little guidance in answering any of them.

*146Under the Court’s reading of § 5, we cannot reach the abridgment question unless we have first determined that a proposed redistricting plan would “lead to a retrogression in the position of racial minorities,” ante, at 141, in comparison to their position under the existing plan. The Court’s conclusion that § 5 demands this preliminary-inquiry is simply wrong; it finds no support in the language of the statute and disserves the legislative purposes behind § 5.

Implicitly admitting as much, the Court adds another question, this one to be asked if the proposed plan is not “retrogressive”: whether “the new apportionment itself so discriminates on the basis of race or color as to violate the Constitution.” Ante, at 141. This addition does much- — in theory, at least — to salvage the Court’s test, since our decisions make clear that the proper test of abridgment under § 5 is essentially the constitutional inquiry.

Still, I cannot accept the Court’s awkward construction. Not only is the Court’s multiple-step inquiry unduly cumbersome and an unnecessary burden to place upon the Attorney General and the District Court for the District of Columbia, but the Court dilutes the meaning of unconstitutionality in this context to the point that the congressional purposes in § 5 are no longer served and the sacred guarantees of the Fourteenth and Fifteenth Amendments emerge badly battered. And in the process, the Court approves a blatantly discriminatory districting plan for the city of New Orleans. I dissent.

I

A

The Fifteenth Amendment provides:

“The right of citizens of the United States to vote shall not be denied or abridged by the United States *147or by any State on account of race, color, or previous condition of servitude.” U. S. Const., Arndt. 15, § 1.

Although the Amendment is self-enforcing, litigation to secure the rights it guarantees proved time consuming and ineffective, while the will of those who resisted its command was strong and unwavering. Finally Congress decided to intervene. In 1965 it enacted the Voting Rights Act, designed “to rid the country of racial discrimination in voting.” South Carolina v. Katzenbach, 383 U. S., at 315. See also id., at 308-315. The Act proclaims that its purpose is “to enforce the fifteenth amendment to the Constitution . . . ,” 79 Stat. 437; the heart of its enforcement mechanism is § 5. In language that tracks that of the Fifteenth Amendment, § 5 declares that no State covered by the Act shall enforce any plan with respect to voting different from that in effect on November 1, 1964, unless the Attorney General or a three-judge District Court in the District of Columbia declares that such plan

“does not have the purpose and will not have the effect of denying or abridging the right to vote on account of race or color . . . .” 42 U. S. C. § 1973c (1970 ed., Supp. V).1

While the substantive reach of § 5 is somewhat broader than that of the Fifteenth Amendment in at least one regard — the burden of proof is shifted from diseriminatee *148to discriminator2 — § 5 is undoubtedly tied to the standards of the Constitution.3 Thus, it is questionable whether the “purpose and effect” language states anything more than the constitutional standard,4 and it is *149clear that the “denying or abridging” phrase does no more- than directly adopt the language of the Fifteenth Amendment.

In justifying its convoluted construction of § 5, however, the Court never deals with the fact that, by its plain language, § 5 does no more than adopt, or arguably expand,5 the constitutional standard. Since it has never *150been held, or even suggested, that the constitutional standard requires an inquiry into whether a redistricting plan is “ameliorative” or “retrogressive,” a fortiori there is no basis for so reading § 5. While the Court attempts to provide a basis by relying on the asserted purpose of § 5 — to preserve present Negro voting strength 6 — it is wholly unsuccessful. What superficial credibility the argument musters is achieved by ignoring not only the statutory language, but also at least three other purposes behind § 5.7

*151Thus, the legislative history of the Voting Rights Act makes clear, and the Court assiduously ignores, that § 5 was designed to preclude new districting plans that “perpetuate discrimination,” 8 to prevent covered jurisdictions *152from “circumventing the guarantees of the 15th amendment” by switching to new, and discriminatory, district-ing plans the moment litigants appear on the verge of having an existing one declared unconstitutional,9 and promptly to end discrimination in voting by pressuring covered jurisdictions to remove all vestiges of discrimination from their enactments before submitting them for preclearance.10 None of these purposes is furthered by an inquiry into whether a proposed districting plan is “ameliorative” or “retrogressive.” Indeed, the statement of these purposes is alone sufficient to demonstrate the error of the Court’s construction.

*153All the purposes of the statute are met, however, by the inquiry § 5’s language plainly contemplates: whether, in absolute terms, the covered jurisdiction can show that its proposed plan meets the constitutional standard. Because it is consistent with both the statutory language and the legislative purposes, this is the proper construction of the provision. Thus, it is the effect of the plan itself, rather than the effect of the change in plans, that should be at issue in a § 5 proceeding.11

Ultimately, the Court admits as much by adding an inquiry into whether the proposed plan, even if “ameliorative,” is constitutional. After this admission, I cannot understand why the Court bothers at all with its preliminary inquiry into the nature of the change of plans, since the inquiry not only adds nothing, but will, I fear, prove to be a time-consuming distraction from the important business of assessing the constitutionality of the proposed plan.12 Except for this unnecessary step, how*154ever, the Court’s final reading of the statute, on its face, no more than duplicates my own.13 Nonetheless, I still do not accept the Court’s approach. After properly re*155turning the constitutional inquiry to the § 5 proceeding, the Court inexplicably tosses off the question in a footnote, and never undertakes the analysis that both our constitutional cases and our § 5 cases have demanded.14 This ultimate denigration of the constitutional standard is a result far short of the promise Congress held out in *156enacting, and re-enacting, the Voting Rights Act, and it is one in which I cannot join.

B

The proper test in § 5 redistricting cases is preordained by our prior cases, which are ignored today by the Court. As suggested above, we have repeatedly recognized the relevance of constitutional standards to the proper construction of § 5. Thus, we have held that in passing that provision “ 'Congress intended to adopt the concept of voting articulated in Reynolds v. Sims, 377 U. S. 533 (1964), and protect Negroes against a dilution of their voting power. ’ ” Perkins v. Matthews, 400 U. S., at 390, quoting Allen v. State Board of Elections, 393 U. S., at 588 (opinion of Harlan, J.). See also Georgia v. United States, 411 U. S., at 532-533; Allen v. State Board of Elections, supra, at 565-566, 569.15 In the Fourteenth Amendment Reynolds line of cases, we have made clear that dilution of voting power refers to resulting voting strength that is something less *157than potential (i. e., proportional) power, not to a reduction of existing power. White v. Regester, 412 U. S. 755, 765-766 (1973); Whitcomb v. Chavis, 403 U. S. 124, 149 (1971). Nonetheless, we have also acknowledged that a showing of less than proportional representation of Negroes by Negro-elected representatives is not alone sufficient to prove unconstitutional dilution:

“To sustain such claims [of dilution], it is not enough that the racial group allegedly discriminated against has not had legislative seats in proportion to its voting potential. The plaintiffs’ burden is to produce evidence to support findings that the political processes leading to nomination and election were not equally open to participation by the group in question — that its members had less opportunity than did other residents in the district to participate in the political processes and to elect legislators of their choice.” White v. Regester, supra, at 765-766.

See also Whitcomb v. Chavis, supra, at 149.16

It is this constitutionally based concept of dilution that we have held to govern in § 5 proceedings. The concept may be readily transferred to the § 5 context simply by adjusting for the shifted burden of proof. Thus, if the proposed redistricting plan underrepresents minority group members, the burden is on the covered *158jurisdiction to show that “the political processes leading to nomination and election were . . . equally open to participation by the group in question.” 17 If the jurisdiction cannot make such a showing, then the proposed plan must be rejected, unless compelling reasons for its adoption can be demonstrated.18

II

• Application of these standards to the case before us is straightforward. Preliminarily, while I agree with the Court that the two at-large seats on the New Orleans City Council are not themselves before the Court for approval and cannot serve as an independent basis for the rejection of Plan II, I do not think Plan II should be assessed without regard to the seven-member council it is designed to fill. Proportional representation of Negroes among the five district seats on the council does not assure Negroes proportional representation on the entire council when, as the District Court found, the two at-large seats will be occupied by white-elected mem*159bers. The Court’s approach of focusing only on the five districts would allow covered municipalities to conceal discriminatory changes by making them a step at a time, and sending one two- or three-district alteration after another to the Attorney General for approval. If nothing beyond the districts actually before him could be considered, discriminatory effects could be camouflaged and the prophylactic purposes of the Act readily evaded.19

Thus the District Court correctly began by considering the seven-member council and a districting plan that, given New Orleans’ long history of racial bloc voting,20 allows Negroes the expectation of no more than one seat (14% of the council), if that, in a city with a 34.5% Negro voting population. Manifestly, the plan serves to underrepresent the Negro voting population. The District Court then, properly, turned to consider whether Negroes are excluded from full participation in the political processes in New Orleans. The court found con*160siderable evidence of both past and present exclusion, none of which is seriously contested here.21

The court found that Louisiana’s majority-vote requirement and “anti-single-shot” requirement operate as a practical matter to defeat Negroes in any district in which they do not constitute a majority,22 that residual effects of Louisiana’s long history of racial discrimination not only in voting, but also in public schools, public assemblies, public recreational facilities, public transportation, housing, and employment, remain; and that city officeholders have generally been unresponsive to the needs of the Negro community. The court looked to the many tactics that, until recently, had been employed with remarkable success to keep Negroes from voting in the State. See Louisiana v. United States, 380 U. S. 145, 147-150 (1965). And the court found that Negro access to the political process is even further narrowed by the fact that candidates in the all-important Democratic primary run on tickets. For a city council candidate to win nomination, which is tantamount to victory in the general election, it is critical to be placed on the ticket of the winning, always white, mayoral candidate. Negro candidates for city council, however, have never been placed on such a ticket. Indeed, no Negro has ever *161been elected to the city council, and the court found that on the rare occasions when a Negro has been elected to any office in the city, it has been because of the support of white candidates or of the white political organization, not because of the power of the Negro electorate. These findings plainly support the District Court’s conclusion that the political processes of New Orleans are not open to Negroes on an equal basis with whites.

Since Negroes are underrepresented by Plan II and have been denied equal access to the political processes in New Orleans, Plan II infringes upon constitutionally protected rights, and only a compelling justification can save the plan. The very nature of the Negro community in New Orleans and the manner of its distortion by Plan II immediately place the city’s explanations in a suspect light. The Negro community is not dispersed, but rather is collected in a concentrated curving band that runs roughly east-west. The districts in Plan II run north-south and divide the Negro community into five parts. Counsel for intervenor Jackson vividly described the effect of this division at oral argument:

“You can walk from Jefferson Parish throughout the city for eight or ten miles through the St. Bernard Parish line and not see a white face along that band, that black belt, that parallels the river in a curve fashion throughout the city. White people live in the very wealthy sections of town out by the lake and along St. Charles Avenue to the river. The rest is left over for blacks, and these are heavy concentrations, and that plan devised by the City Council slices up that population like so many pieces of bologna . . . .” Tr. of Oral Arg. 30.

As Jonathan A. Eckert, the council staff member pri*162marily responsible for drafting Plan II, conceded in the District Court, the “inevitable result” of Plan IPs north-south orientation is “to have districts in which blacks are generally in the minority, or at the most in a bare majority.” 2 App. 346.

New Orleans relies on seven goals that it claims mandate a north-south scheme such as Plan II. The city’s own belief in this conclusion is questionable in light of Mr. Eckert’s testimony in the District Court that he and his staff had drafted at least two east-west plans that satisfied them. 1 App. 336-337. In any case, however, the asserted goals, whether taken alone or in combination, do not establish a compelling justification for the plan. One claimed purpose is to prevent dilution of the vote of minority groups. Plan II plainly does not achieve this goal. Two other asserted aims are to achieve substantial numerical equality among the five districts and to keep the resultant districts compact and contiguous. Both aims can be accomplished by any number of east-west plans as well. Three more proffered justifications are to preserve ward and precinct lines, natural boundaries, and manmade boundaries. But there are findings that ward lines cannot be observed in any case because of one-person, one-vote restrictions, and that precincts are sufficiently small that their integrity can be honored in east-west districts. This latter fact minimizes any adverse effects of violating natural and manmade boundaries, except to the extent that they divide communities of different social or economic interests. And Plan II only erratically keeps such communities intact.

It is only the seventh of the proffered goals that, if compelling, mandates a north-south scheme: keeping incumbents apart in the new districts so that they will *163not have to run against one another for re-election.23 Four of the five district councilmen live in an east-west line along the lake in the northern part of the city. East-west districts would place all four in the same one or two districts, 1 App. 125, 232, 235, and north-south lines are therefore necessary if these councilmen are to remain apart. 2 App. 344. While the desire to keep incumbents in separate districts may have merit in some contexts, it surely cannot stand alone to justify the substantial dilution of minority voting rights found here.

Thus, the city has failed to show an acceptable justification for the racially dilutive effect of Plan II. Accordingly, the District Court correctly concluded that appellants failed to demonstrate that Plan II would not have the effect of abridging the right to vote on account of race, and correctly denied the requested declaratory judgment.24